Opportunities for prevention of 'clinically significant' knee pain: Results from a population-based cross sectional survey

Background: There is little UK-based evidence on the prevalence and predictors of knee pain associated with disability across all adult ages. We aimed to estimate the prevalence of 'clinically significant' knee pain, identify and assess the population impact of independent risk factors, and estimate levels of healthcare need. Methods: A cross-sectional survey of three general practice populations was conducted. Adults (n = 5752) were mailed a screening questionnaire (phase I). Those reporting predominant or isolated knee pain were sent a detailed questionnaire (phase II), with a further sub-sample invited for clinical examination (phase III). Logistic regression was used to identify independent risk factors and population attributable fractions were calculated. Results: The 1 month period prevalence of 'all reported' knee pain was 19 per cent, of which about a third was disabling and a fifth intense and disabling. Obesity, deprivation and South Asian ethnicity were each associated with a 3-4-fold increased risk of knee pain with disability (after age/sex adjustment). The attributable fraction estimate for raised body mass index (BMI) was 36 per cent (27-44 per cent) - the population impact of being overweight was greater than that of being obese. Thirteen per cent of all adults reported a previous primary care consultation, 7 per cent reported previous secondary care referral for knee pain, and 4.5 per cent (2.7-6.2 per cent) of the adult population were currently receiving or in need of specialist treatment. Conclusions: The high population impact of being overweight (BMI 25-29) or obese (BMI 30 or more) has implications for primary prevention. The estimates of previous healthcare usage, and of levels of met and unmet need, are useful for healthcare planning.